What are you testing for?
A benchmark provides a reference point for comparing hardware and software. With VRMark benchmarks, we’re simulating real-world VR content to test a system’s performance, latency and accuracy.
- The performance requirements for VR games are much higher than for typical PC games. Both the display resolution and frame rate need to be higher to deliver a comfortable VR experience. Running VRMark benchmarks tells you whether a PC has the necessary performance to meet those demands.
- Latency is the time needed for the movement of a VR headset to be fully perceived on the headset’s display. Low latency is essential to create the feeling of presence, the feeling that you are actually in the virtual world. High latency results in a poor experience. When the screen lags behind the user’s movement, it can result in disorientation, motion sickness, and nausea.
- Accuracy testing checks whether the intended frames are shown on the VR headset and whether the system is correctly tracking user movement and input. VR systems use clever techniques to predict movement and compensate for missed frames, but poor accuracy may leave some users feeling uncomfortable and distracted.
VRMark looks at the complete VR system. The benchmarks, which can be run with or without a VR headset, focus on measuring processing performance, while using VRMark’s Experience Mode with a headset lets you evaluate the quality of the VR experience.
How do you do it?
We are committed to creating industry standard benchmarks that are relevant, accurate and impartial. Our benchmarks are developed in partnership with the world’s leading technology companies. We have a transparent development process that’s been government vetted for fairness and neutrality.
The development process for VRMark began by consulting widely within the industry to determine the needs and requirements for a VR benchmark. The benchmark workloads were then carefully crafted and calibrated to reflect the demands of real VR games and experiences.
We use the same tools that VR developers use, but by keeping development in-house and maintaining strict control over the source code, we ensure that our benchmarks are fair and free from bias.
What do you find among systems with lower scores?
We find systems with lower performance struggle to meet the target frame rate and may exhibit greater latency. Poorly performing hardware can leave users feeling nauseated and dizzy during the VR experience. Such systems are easily identified by their low scores in VRMark tests.
One surprise during development, however, was just how well VR systems can compensate for missed frames. Some incredibly bright people are tackling VR’s performance challenges in very creative ways. As a result, even when the frame rate is low, many users are surprised by the quality of the VR experience.
Who can download the software?
VRMark is available in a range of editions to meet the needs of different user groups. The software is free for consumers and the press. Consumers can test their home PCs to ensure they have enough processing power for VR before buying a headset. The press uses our benchmarks to complement their reviews with objective performance measurements and comparisons.
Manufactures and other commercial users pay a licensing fee to use VRMark Professional Edition. Manufacturers commonly use benchmarks for product testing, research and development, component evaluation, competitor analysis and sales and marketing activities.
Are benchmark results available to the public?
Yes! We maintain the world’s largest and most comprehensive hardware performance database, using the results submitted by millions of users to help consumers make informed purchasing decisions. VRMark benchmark scores are available for most commercially available products.
Do you test for AR or Mixed Reality?
Currently, VRMark is focused on testing the performance of virtual reality systems. In the future, we hope to create benchmark tests for AR and mixed reality systems as well.